How to using euler-sr to assebling 454 seq?

[kentnf]

I install eular in ubuntu8.10 x64, it runs well assembling sample data.
command is : ${EUSRC}/assembly/Assemble.pl reads.fa 25

Then i use my 454 seqs, it has been clean ahead.
command is: ${EUSRC}/assembly/Assemble.pl 454reads.fa 25

But the result looks have some mistakes. two assemblied seqs like that:
>RUX0000005
TGAATGTGTTCATGATGGTTGGGGGGTTGCTGTACTTGTTGGAGTG
>RUX0000006
TGAATGTGTTCATGATGGTTGGGGGTTGCTGTACTTGTTGGAGTG

I think they are the same, but they are not assembly to one seq using euler-sr.

Who can tell me what the problem I have?

Thank you very much.

PS: I also want to know how to check the quality of the contig file like cap3 or phrap results.

[mchaisso]

Most sequences like this get merged, so I'm not sure what happened in this case.

You can manually re-do some of the steps in the assembly to get rid of these with the following commands:

mkdir simple2
${EUSRC}/assembly/${MACHTYPE}/simplifyGraph simple/454reads.fa simple2/454reads.fa -removeBulges 75 -removeWhirls 75
${EUSRC}/assembly/${MACHTYPE}/transformGraph simple2/454reads.fa transformed/454reads.fa -minPathCount 3 -notStrict
${EUSRC}/assembly/${MACHTYPE}/printContigs transformed/454reads.fa
cp transformed/reads.fa.contig .

I've added a few changes in the forthcoming release that improve assembly results on unpaired 454 reads (or general improvements in the step that resolves repeats without mate pairs). If you want to post these reads I'll try and make sure this assembly is fixed in the release.

-mark